1. Field of the Invention
The present invention relates to a technique of displaying on a display, photographing, and recognizing an image into which data is embedded in such a way that the embedded data cannot easily be seen by human eyes.
2. Description of the Related Art
As a conventional technique of embedding data into an image in such a way that it cannot easily be seen by human eyes, the technique disclosed in Patent Document 1 is known.
In this conventional technique, a yellow component of an original image is divided into blocks of a predetermined size, and the data to which error correction data is added is repeatedly embedded into a pair of blocks on the basis of a magnitude relation of the difference in average density of adjacent blocks of the block pair.
Data acquisition from the image in which data is embedded can be realized by, firstly, obtaining the data that is repeatedly embedded on the basis of a magnitude relation of the average density of adjacent blocks of the block pair (original data to which error correction data is added), and then by performing majority decision and error correction. Performing majority decision allows the number of errors in the data to be reduced, and performing error correction allows the original data to be obtained with errors in the data corrected.
As majority decision and error correction enable to obtain accurate data, embedded data can also be obtained from images captured by image input means such as a camera, with the image containing embedded data being printed on a piece of paper or displayed on a screen.
[Patent Document 1] Japanese Patent Application Publication No. 2004-349879
However, an image obtained via image output means such as print or a display and image input means such as a camera is deteriorated from the original image. Therefore, in the above-mentioned conventional data embedding method, processes such as providing data for error correction or repeatedly embedding the same data are required in order to obtain data correctly from the images. Thus there has been a problem whereby the amount of data that is actually embeddable is limited, and this amount has been difficult to increase.